__dl_thread_cleanup is called from the context of an exiting thread
that is not in a consistent state valid for calling application code.
since commit c9f415d7ea, it's possible
(and supported usage) for the allocator to have been replaced by the
application, so __dl_thread_cleanup can no longer call free. instead,
reuse the message buffer as a linked-list pointer, and queue it to be
freed the next time any dynamic linker error message is generated.
the way gets was implemented in terms of fgets, it used the location
of the null termination to determine where to find and remove the
newline, if any. an embedded null byte prevented this from working.
this also fixes a one-byte buffer overflow, whereby when gets read an
N-byte line (not counting newline), it would store two null
terminators for a total of N+2 bytes. it's unlikely that anyone would
care that a function whose use is pretty much inherently a buffer
overflow writes too much, but it could break the only possible correct
uses of this function, in conjunction with input of known format from
a trusted/same-privilege-domain source, where the buffer length may
have been selected to exactly match a line length contract.
there seems to be no correct way to implement gets in terms of a
single call to fgets or scanf, and using multiple calls would require
explicit locking, so we might as well just write the logic out
explicitly character-at-a-time. this isn't fast, but nobody cares if a
catastrophically unsafe function that's so bad it was removed from the
C language is fast.
in order to implement ENOTRECOVERABLE, the implementation has
traditionally used a bit of the mutex type field to indicate that it's
recovered after EOWNERDEAD and will go into ENOTRECOVERABLE state if
pthread_mutex_consistent is not called before unlocking. while it's
only the thread that holds the lock that needs access to this
information (except possibly for the sake of pthread_mutex_consistent
choosing between EINVAL and EPERM for erroneous calls), the change to
the type field is formally a data race with all other threads that
perform any operation on the mutex. no individual bits race, and no
write races are possible, so things are "okay" in some sense, but it's
still not good.
this patch moves the recovery/consistency state to the mutex
owner/lock field which is rightfully mutable. bit 30, the same bit the
kernel uses with a zero owner to indicate that the previous owner died
holding the lock, is now used with a nonzero owner to indicate that
the mutex is held but has not yet been marked consistent. note that
the kernel ABI also reserves bit 29 not to appear in any tid, so the
sentinel value we use for ENOTRECOVERABLE, 0x7fffffff, does not clash
with any tid plus bit 30.
fdopendir is specified to fail with EBADF if the file descriptor
passed is not open for reading. while O_PATH is an extension and
arguably exempt from this requirement, it's used, albeit incompletely,
to implement O_SEARCH, and fdopendir should fail when passed an
O_SEARCH file descriptor.
the new check is performed after fstat so that we don't have to
consider the possibility that the fd is invalid.
an alternate solution would be attempting to pre-fill the buffer using
getdents, which would fail with EBADF for us, but that seems more
complex and error-prone and involves either code duplication or
refactoring, so the simple fix with an additional inexpensive syscall
is what I've made for now.
Some packages call gettext to format a message to be sent to perror.
If the currently set user locale points to a non-existent .mo file,
open via __map_file in dcngettext will set errno to ENOENT.
Maintainer's notes: Non-modification of errno is a documented part of
the interface contract for the GNU version of this function and likely
other versions. The issue being fixed here seems to be a regression
from commit 1b52863e24, which enabled
setting of errno from __map_file.
commit 84d061d5a3 attempted to do this
already, but omitted from pthread_key_create.c the weak definition of
__pthread_key_delete_synccall, so that the definition provided by
pthread_key_delete.c was always pulled in.
based on patch by Markus Wichmann, but with a weak alias rather than
weak reference for consistency/policy about dependence on tooling
features.
fallback to /etc/shadow should happen only when the entry is not found
in the TCB shadow. otherwise transient errors or permission errors can
cause inconsistent results.
this reverts commit c0ed5a201b, which
was based on a mistaken reading of POSIX due to inconsistency between
the description (which requires return upon interruption by a signal)
and the errors list (which wrongly lists EINTR as "may fail").
since the previously-introduced behavior was a workaround for an old
kernel bug to ensure safety of correct programs that were not hardened
against the bug, an effort has been made to preserve it for programs
which do not use interrupting signal handlers. the stage for this was
set in commit a63c0104e4, which makes
the futex __timedwait backend suppress EINTR if it's seen when no
interrupting signal handlers have been installed.
based loosely on a patch submitted by Orivej Desh, but with
unnecessary additional changes removed.
the resolution of Austin Group issue #1132 changes the requirement to
fail so that it only applies when the set argument (new mask) is
non-null. this change was made for consistency with the description,
which specified "if set is a null pointer, the value of the argument
how is not significant".
prior to linux 2.6.22, futex wait could fail with EINTR even for
non-interrupting (SA_RESTART) signals. this was no problem provided
the caller simply restarted the wait, but sem_[timed]wait is required
by POSIX to return when interrupted by a signal. commit
a113434cd6 introduced this behavior, and
commit c0ed5a201b reverted it based on a
mistaken belief that it was not required. this belief stems from a bug
in the specification: the description requires the function to return
when interrupted, but the errors section marks EINTR as a "may fail"
condition rather than a "shall fail" one.
since there does seem to be significant value in the change made in
commit c0ed5a201b, making it so that
programs that call sem_wait without checking for EINTR don't silently
make forward progress without obtaining the semaphore or treat it as a
fatal error and abort, add a behind-the-scenes mechanism in the
__timedwait backend to suppress EINTR in programs that have never
installed interrupting signal handlers, and have sigaction track and
report this state. this way the semaphore code is not cluttered by
workarounds and can be updated (to be done in next commit) to reflect
the high-level logic for conforming behavior.
these changes are based loosely on a patch by Markus Wichmann, with
the main changes being atomic update to flag object and moving the
workaround from sem_timedwait to the __timedwait futex backend.
it's not clear whether this is required, but it seems arguable that it
should happen. for example aio_suspend is supposed to return
immediately if any of the operations has "completed", which includes
ending with an error status asynchonously and might also be
interpreted to include doing so synchronously.
the map structures in particular are permanent once created, and thus
a large number of aio function calls with invalid file descriptors
could exhaust memory, whereas, assuming normal resource limits, only a
very small number of entries ever need to be allocated. check validity
of the fd before allocating anything new, so that allocation of large
amounts of memory is only possible when resource limits have been
increased and a large number of files are actually open.
this change also improves error reporting for bad file descriptors to
happen at the time the aio submission call is made, as opposed to
asynchronously.
since commit c9f415d7ea, it has been
possible that the allocator is application-provided code, which cannot
necessarily run safely on io thread stacks, and which should not be
able to see the existence of io threads, since they are an
implementation detail.
instead of having the io thread request and possibly allocate its
queue (and the map structures leading to it), make the submitting
thread responsible for this, and pass the queue pointer into the io
thread via its args structure. this eliminates the only early error
case in io threads, making it no longer necessary to pass an error
status back to the submitting thread via the args structure.
aio threads not using SIGEV_THREAD notification are created with small
stacks and no guard page, which is possible since they only run the
code for the requested io operation, not any application code. the
motivation is not creating a lot of VMAs. however, the io thread needs
to be able to receive a cancellation signal in case aio_cancel
(implemented via pthread_cancel) is called. this requires sufficient
stack space for a signal frame, which PTHREAD_STACK_MIN does not
necessarily include.
in principle MINSIGSTKSZ from signal.h should give us sufficient space
for a signal frame, but the value is incorrect on some existing archs
due to kernel addition of new vector register support without
consideration for impact on ABI. some powerpc models exceed
MINSIGSTKSZ by about 0.5k, and x86[_64] with AVX-512 can exceed it by
up to about 1.5k. so use MINSIGSTKSZ+2048 to allow for the discrepancy
plus some working space.
unfortunately, it's possible that signal frame sizes could continue to
grow, and some archs (aarch64) explicitly specify that they may.
passing of a runtime value for MINSIGSTKSZ via AT_MINSIGSTKSZ in the
aux vector was added to aarch64 linux, and presumably other archs will
use this mechanism to report if they further increase the signal frame
size. when AT_MINSIGSTKSZ is present, assume it's correct, so that we
only need a small amount of working space in addition to it; in this
case just add 512.
first, the condition (mem && k < p) is redundant, because mem being
nonzero implies the needle is periodic with period exactly p, in which
case any byte that appears in the needle must appear in the last p
bytes of the needle, bounding the shift (k) by p.
second, the whole point of replacing the shift k by mem (=l-p) is to
prevent shifting by less than mem when discarding the memory on shift,
in which case linear time could not be guaranteed. but as written, the
check also replaced shifts greater than mem by mem, reducing the
benefit of the shift. there is no possible benefit to this reduction of
the shift; since mem is being cleared, the full shift is valid and
more optimal. so only replace the shift by mem when it would be less
than mem.
commit ddc947eda3 fixed the
corresponding bug for exit which was introduced when commit
0b80a7b040 added support for
caller-provided buffers, making it possible for stderr to be a
buffered stream.
fflush(NULL) and __stdio_exit lock individual FILEs while holding the
open file list lock to walk the list. since fclose first locked the
FILE to be closed, then the ofl lock, it could deadlock with these
functions.
also, because fclose removed the FILE to be closed from the open file
list before flushing and closing it, a concurrent fclose or exit could
complete successfully before fclose flushed the FILE it was closing,
resulting in data loss.
reorder the body of fclose to first flush and close the file, then
remove it from the open file list only after unlocking it. this
creates a window where consumers of the open file list can see dead
FILE objects, but in the absence of undefined behavior on the part of
the application, such objects will be in an inactive-buffer state and
processing them will have no side effects.
__unlist_locked_file is also moved so that it's performed only for
non-permanent files. this change is not necessary, but preserves
consistency (and thereby provides safety/hardening) in the case where
an application uses one of the standard streams after closing it while
holding an explicit lock on it. such usage is of course undefined
behavior.
Use "+r" in the asm instead of implementing a non-transparent copy by
applying "0" constraint to the source value. Introduce a typedef for
the function type to avoid spelling it out twice.
commit aeeac9ca54 introduced fail-safe
invariants that creating a locale_t object for the C locale or C.UTF-8
locale will always succeed. extend the guarantee to also cover the
following:
- newlocale(LC_ALL_MASK, "", 0)
- newlocale(LC_ALL_MASK-LC_CTYPE_MASK, "C", 0)
provided that the LANG/LC_* environment variables have not been
changed by the program. these usages are idiomatic for getting the
default locale, and for getting a locale that behaves as the C locale
except for honoring the default locale's character encoding.
unify the code paths for allocated and non-allocated locale objects,
always using a tmp object. this is necessary to avoid clobbering the
base locale object too soon if we allow for the possibility that
looking up an explicitly requested locale name may fail, and makes the
code simpler and cleaner anyway.
eliminate the complex and fragile logic for checking whether one of
the non-allocated locale objects can be used for the result, and
instead just memcmp against each of them.
commit 63c188ec42 missed making this
change when switching from atomics to locking for modification of the
global locale, leaving access to locale structures unnecessarily
burdened with the restrictions of volatile.
the volatile qualification was originally added in commit
56fbaa3bbe.
introduce a new LOC_MAP_FAILED sentinel for errors, since null
pointers for a category's locale map indicate the C locale. at this
time, __get_locale does not fail, so there should be no functional
change by this commit.
the choice of signed char for lbf was a theoretically space-saving
hack that was not helping, and was unwantedly expensive. while
comparing bytes against a byte-sized member sounds easy, the trick
here was that the byte to be compared was unsigned while the lbf
member was signed, making it possible to set lbf negative to disable
line buffering. however, this imposed a requirement to promote both
operands, zero-extending one and sign-extending the other, in order to
compare them.
to fix this, repurpose the waiters count slot (unused since commit
c21f750727). while we're at it, switch
mode (orientation) from signed char to int as well. this makes no
semantic difference (its only possible values are -1, 0, and 1) but it
might help on archs where byte access is awkward.
to check whether flush due to line buffering is needed, the int-type
character argument must be truncated to unsigned char for comparison.
if the original value is subsequently passed to __overflow, it must be
preserved, adding to register pressure. since it doesn't matter,
truncate all uses so the original value is no longer live.
the internal putc_unlocked macro was wrongly returning a meaningless
boolean result rather than the written character or EOF.
bug was found by reading (very surprising) asm.
check whether the lock is free before loading the calling thread's
tid. if so, just use a dummy tid value that cannot compare equal to
any actual thread id (because it's one bit wider). this also avoids
the need to save the tid and pass it to locking_getc or locking_putc,
reducing register pressure.
this change might slightly hurt the case where the caller already
holds the lock, but it does not affect the single-threaded case, and
may significantly improve the multi-threaded case, especially on archs
where loading the thread pointer is disproportionately expensive like
early mips and arm ISA levels. but even on i386 it helps, at least on
some machines; I measured roughly a 10-15% improvement.
this is not needed for correctness, but doesn't hurt, and in some
cases the compiler may pessimize the call assuming the callee might be
variadic when it lacks a prototype.
commit 4390383b32 inadvertently used "r"
instead of "0" for the input constraint, which only happened to work
for the configuration I tested it on because it usually makes sense
for the compiler to choose the same input and output register.
by ABI, the public stdin/out/err macros use extern pointer objects,
and this is necessary to avoid copy relocations that would be
expensive and make the size of the FILE structure part of the ABI.
however, internally it makes sense to access the underlying FILE
objects directly. this avoids both an indirection through the GOT to
find the address of the stdin/out/err pointer objects (which can't be
computed PC-relative because they may have been moved to the main
program by copy relocations) and an indirection through the resulting
pointer object.
in most places this is just a minor optimization, but in the case of
getchar and putchar (and the unlocked versions thereof), ipa constant
propagation makes all accesses to members of stdin/out PC-relative or
GOT-relative, possibly reducing register pressure as well.
with these changes, in a program that has not created any threads
besides the main thread and that has not called f[try]lockfile, getc
performs indistinguishably from getc_unlocked. this was measured on
several i386 and x86_64 models, and should hold on other archs too
simply by the properties of the code generation.
the case where the caller already holds the lock (via flockfile) is
improved significantly as well (40-60% reduction in time on machines
tested) and the case where locking is needed is improved somewhat
(roughly 10%).
the key technique used here is forcing the non-hot path out-of-line
and enabling it to be a tail call. a static noinline function
(conditional on __GNUC__) is used rather than the extern hiddens used
elsewhere for this purpose, so that the compiler can choose
non-default calling conventions, making it possible to tail-call to a
callee that takes more arguments than the caller on archs where
arguments are passed on the stack or must have space reserved on the
stack for spilling the. the tid could just be reloaded via the thread
pointer in locking_getc, but that would be ridiculously expensive on
some archs where thread pointer load requires a trap or syscall.
on multiple occasions I've started to flatten/inline the code in
__init_libc, only to rediscover the reason it was not inlined: GCC
fails to deallocate its stack (and now, with the changes in commit
4390383b32, fails to produce a tail call
to the stage 2 function; see PR #87639) before calling main if it was
inlined.
document this with a comment and use an explicit noinline attribute if
__GNUC__ is defined so that even with CFLAGS that heavily favor
inlining it won't get inlined.
this is the analog of commit 1c84c99913
for static linking. unlike with dynamic linking, we don't have
symbolic lookup to use as a barrier. use a dummy (target-agnostic)
degenerate inline asm fragment instead. this technique has precedent
in commit 05ac345f89 where it's used for
explicit_bzero. if it proves problematic in any way, loading the
address of the stage 2 function from a pointer object whose address
leaks to kernelspace during thread pointer init could be used as an
even stronger barrier.
fma is only available on recent x86_64 cpus and it is much faster than
a software fma, so this should be done with a runtime check, however
that requires more changes, this patch just adds the code so it can be
tested when musl is compiled with -mfma or -mfma4.
vfma is available in the vfpv4 fpu and above, the ACLE standard feature
test for double precision hardware fma support is
__ARM_FEATURE_FMA && __ARM_FP&8
we need further checks to work around clang bugs (fixed in clang >=7.0)
&& !__SOFTFP__
because __ARM_FP is defined even with -mfloat-abi=soft
&& !BROKEN_VFP_ASM
to disable the single precision code when inline asm handling is broken.
For runtime selection the HWCAP_ARM_VFPv4 hwcap flag can be used, but
that requires further work.
previously (before and after rewrite), spurious escaping of path
separators as \/ was not treated the same as /, but rather got split
as an unpaired \ at the end of the fnmatch pattern and an unescaped /,
resulting in a mismatch/error.
for the case of \/ as part of the maximal literal prefix, remove the
explicit rejection of it and move the handling of / below escape
processing.
for the case of \/ after a proper glob pattern, it's hard to parse the
pattern, so don't. instead cheat and count repetitions of \ prior to
the already-found / character. if there are an odd number, the last is
escaping the /, so back up the split position by one. now the
char clobbered by null termination is variable, so save it and restore
as needed.
this code has been long overdue for a rewrite, but the immediate cause
that necessitated it was total failure to see past unreadable path
components. for example, A/B/* would fail to match anything, even
though it should succeed, when both A and A/B are searchable but only
A/B is readable. this problem both was caught in conformance testing,
and impacted users.
the old glob implementation insisted on searching the listing of each
path component for a match, even if the next component was a literal.
it also used considerable stack space, up to length of the pattern,
per recursion level, and relied on an artificial bound of the pattern
length by PATH_MAX, which was incorrect because a pattern can be much
longer than PATH_MAX while having matches shorter (for example, with
necessarily long bracket expressions, or with redundancy).
in the new implementation, each level of recursion starts by consuming
the maximal literal (possibly escaped-literal) path prefix remaining
in the pattern, and only opening a directory to read when there is a
proper glob pattern in the next path component. it then recurses into
each matching entry. the top-level glob function provided automatic
storage (up to PATH_MAX) for construction of candidate/result strings,
and allocates a duplicate of the pattern that can be modified in-place
with temporary null-termination to pass to fnmatch. this allocation is
not a big deal since glob already has to perform allocation, and has
to link free to clean up if it experiences an allocation failure or
other error after some results have already been allocated.
care is taken to use the d_type field from iterated dirents when
possible; stat is called only when there are literal path components
past the last proper-glob component, or when needed to disambiguate
symlinks for the purpose of GLOB_MARK.
one peculiarity with the new implementation is the manner in which the
error handling callback will be called. if attempting to match */B/C/D
where a directory A exists that is inaccessible, the error reported
will be a stat error for A/B/C/D rather than (previous and wrong
implementation) an opendir error for A, or (likely on other
implementations) a stat error for A/B. such behavior does not seem to
be non-conforming, but if it turns out to be undesirable for any
reason, backtracking could be done on error to report the first
component producing it.
also, redundant slashes are no longer normalized, but preserved as
they appear in the pattern; this is probably more correct, and falls
out naturally from the algorithm used. since trailing slashes (which
force all matches to be directories) are preserved as well, the
behavior of GLOB_MARK has been adjusted not to append an additional
slash to results that already end in slash.
Rich Felker
A. Wilcox
Quentin Rameau
Alexander Monakov
Szabolcs Nagy